US 7272129 B2
A method and apparatus for synchronizing a node (200) within an ad-hoc communication system (100) is described herein. During operation all nodes periodically broadcast a synchronization beacon for other nodes to utilize for synchronization when a coordinating access point (node) is unavailable. A particular node's synchronization beacon will have an associated “tier” number that is incremented from the tier number of the beacon used to synchronize the particular node. In the absence of an access point, a node that joins the ad-hoc communication system will listen for synchronization beacons transmitted by other nodes. If synchronization beacons are heard, the node will synchronize with a beacon having a lowest tier. The node will then broadcast its own beacon having its tier number incremented from the lowest tier beacon heard.
1. A method for a node to synchronize to an ad-hoc communication system, the method comprising the steps of:
listening for a plurality of synchronization beacons transmitted from a plurality of nodes;
if synchronization beacons are heard, performing the steps of:
determining a tier for each synchronization beacon heard;
synchronizing to a beacon having a lowest tier;
transmitting a beacon having a tier greater than the lowest tier; and
if synchronization beacons are not heard, performing the steps of:
transmitting a beacon having a first tier.
2. The method of
creating the beacon comprising a beacon identification and a tier number; and
transmitting the beacon having the beacon identification and a tier number.
3. The method of
4. The method of
5. The method of
6. The method of
7. The method of
determining that a beacon exists having a lower tier than the beacon utilized for synchronization; and
reestablishing synchronization to the beacon having the lower tier.
8. The method of
transmitting a beacon having a tier greater than the beacon used to synchronize with.
The present invention relates generally to ad-hoc communication systems and in particular, to a method and apparatus for synchronizing a node within such ad-hoc communication systems.
Synchronization of nodes within ad-hoc communication systems is critical to proper system performance. Synchronization of nodes requires that each node's internal clock be set to the same system time within some margin of error. When nodes are synchronized, power-saving techniques can be implemented. Particularly, nodes can power down (sleep) for a predetermined period of time and power up (wake) at a specified time to insure that messages can be exchanged. Therefore, a need exists for a method and apparatus for synchronizing a node within an ad-hoc communication system.
In order to address the above-mentioned need, a method and apparatus for synchronizing a node within an ad-hoc communication system is described herein. During operation, nodes periodically broadcast a synchronization beacon for other nodes to utilize for synchronization when a coordinating access point is unavailable. A particular node's synchronization beacon will have an associated “tier” number that is incremented from the tier number of the beacons used to synchronize the particular node. In the absence of an access point, a node that joins the ad-hoc communication system will listen for synchronization beacons transmitted by other nodes. If synchronization beacons are heard, the node will synchronize with beacons having a lowest tier. A clock bias/offset is adjusted based on receiving one or more beacons. Once the node is synchronized, the node will then broadcast its own beacon having its tier number incremented from the lowest tier beacon heard.
The above synchronization technique will allow for finer synchronization of nodes over a multiple hop network. The solution maintains compatibility to previously defined power save operation and does not require any modifications or indications from the physical layer. Additionally, the above synchronization technique uses minimal signaling overhead by utilizing existing messaging, and does not require any complex computation or processing on the node.
The present invention encompasses a method for a node to synchronize to an ad-hoc communication system. The method comprises the steps of listening for a plurality of synchronization beacons transmitted from a plurality of nodes. If synchronization beacons are heard, then a tier for each synchronization beacon heard is determined and synchronization takes place to a beacon having a lowest tier. A beacon is then transmitted having a tier greater than the lowest tier. If, however synchronization beacons are not heard, then a beacon is transmitted having a first tier.
The present invention additionally encompasses a method comprising the steps of determining that a synchronization beacon cannot be heard, creating a beacon having a tier number equal to zero, and a beacon identification field based on a node's physical address, and transmitting the beacon having the tier number and beacon identification field.
The present invention additionally encompasses an apparatus comprising a receiver listening for a plurality of synchronization beacons transmitted from a plurality of nodes, logic circuitry determining a tier for each synchronization beacon heard and synchronizing to a beacon having a lowest tier, and a transmitter transmitting a beacon having a tier greater than the lowest tier.
Turning now to the drawings, wherein like numerals designate like components,
If a coordinating node beacon cannot be received, logic circuitry 201 accesses receiver 205 and determines if synchronization beacons from other non-coordinating nodes 20 can be received. If other nodes synchronization beacons are received, the beacons' tiers are analyzed and synchronization takes place (as described above) utilizing the beacons having the lowest tier. Beacon parameters are then updated by logic circuitry 201. Such beacon parameters include, but are not limited to:
If, however, no coordinating device is heard, and after a predetermined period of time (Tpresync) no beacon is heard from non-coordinating nodes, node 200 will become a first tier (tier #0) node. Beacon parameters are updated accordingly by logic circuitry 201 to reflect this. Logic circuitry 201 will then instruct receiver 205 to periodically transmit the tier #0 beacon with a beacon ID based upon the physical address of the tier #0 node. (A physical address is a hardware address that uniquely identifies each node of a network and is unchanging. Such an address is usually “hard wired” into the node during its manufacture. In networks using an IEEE 802.11 protocol, the physical address comprises a Medium Access Controller (MAC) address).
No matter what tier # a node's beacon currently is associated with, all nodes in the network share the responsibility of periodically broadcasting beacons at beacon intervals relative to the adjusted clock. If a neighboring node hears the beacons, it will synchronize to the beacon having the lowest tier #, and begin transmitting its own synchronization beacon with a beacon ID based upon the Medium Access Controller Address of the tier #0 node.
If, at step 301, no coordinating node's beacon is heard, the logic flow continues to step 307 where logic circuitry 201 accesses receiver 205 and determines if beacons from non-coordinating nodes are heard. If so, the logic flow continues to step 309 where logic circuitry 201 determines a tier for each beacon heard and synchronizes to the beacon(s) having the lowest tier (TIERLOWEST). The logic flow then continues to step 311 where logic circuitry 201 updates beacon parameters to become a tier #X beacon, where X is a numeral incremented by one from TIERLOWEST. The logic flow then continues to step 315.
Returning to step 307, if logic circuitry 201 determines that no synchronization beacons are heard, then the logic flow continues to step 313 where logic circuitry updates beacon parameters to become a tier #0 node (e.g., first tier, or Tier Number=0 node). The logic flow continues to step 315.
At step 315 logic circuitry creates a synchronization beacon comprising a beacon identification and the tier number and instructs transmitter 203 to periodically broadcast the synchronization beacon. As discussed above, the beacon identification comprise a physical address (e.g., a MAC address) of the tier #0 node.
In a synchronization maintenance mode, each node 200 continuously monitors the Beacon Transmission Time received in beacons from neighboring nodes to determine if a beacon exists having a lower tier than the beacon utilized for synchronization. Once beacons are identified containing a lower tier number, node 200 begins to reestablish synchronization with these beacons. When Nsync
In synchronization maintenance mode, all nodes in the network share responsibility to send beacons relative to the adjusted clock. Additionally, all nodes shall rebroadcast beacons with the same BSSID, Beacon ID, Beacon Interval, and Infrastructure Access Index indication values that were received in beacons which node 200 utilized for synchronization. Additionally, all nodes randomly set a timeout to force a resynchronization of the network after at least Tresync
The resynchronization procedure ensures that the node's reference clock 207 is maintained and that the tier #0 node that all other nodes synchronize to is periodically changing. This is particularly useful if node 200 has disconnected or is moving away from the synchronized network. Resynchronization is performed periodically (after a timeout equal to Tresync
During resynchronization, a new beacon ID (based on the self appointed node's MAC address) is broadcast in the self appointed node's beacons. All nodes that receive a beacon with a new beacon ID and the same BSSID (regardless of the tier number they are currently affiliated with) will recognize the beacon as a resynchronization procedure.
When each node receives the resynchronization beacon, it shall add a backoff timer of Tsync
As synchronized nodes move about, a merge of independently synchronized groups of nodes with different BSSIDs may be necessary as the groups come together. This merge will enable the two networks to use a common reference clock.
A base station ID (BSSID) uniquely identifies a network of synchronized nodes. Within a synchronized network, resynchronization will alter the beacon ID. There is a need to also uniquely identify a network so that when two networks come together, the need to merge is recognized. Without this, if a node saw a different beacon ID, it would think this is a resynchronization rather than a merging of networks. To provide a consistent approach for a merge of these networks, the BSSID is used to determine which network will maintain the reference synchronization clock and which must resynchronize to the reference clock. The method for detecting the need for a merge is triggered by the reception by a node of a beacon with a BSSID that is different than the one it currently is synchronized with. Upon recognizing the presence of another network with a different BSSID, node 200 will compare both the Infrastructure Access Index indication and the BSSID in the beacon received from the other network with its own copy of these parameters to determine if a network merge is required.
If a network merge is required, then node 200 will set the Network Merge indication in its beacon for the next Nmerge
After broadcasting Nmerge
In an alternate embodiment, the requirement for a network merge may be assessed based upon the relative ad-hoc network size rather than the absolute value of the BSSID. Thus, when the presence of a different network is detected based on the BSSID of a received beacon, an optional beacon parameter (Relative Network Size) containing a representation of the relative size of the adjacent network is compared with the receiving node's estimation of the network size. The Relative Network Size is a bit map where each bit is a hashed index of an ad hoc network node's MAC address. The number of bits is engineered for desired accuracy. In this alternative embodiment, the number of bits that are set represents the size of the network. Alternatively, the actual network size may be computed and used as the Relative Network Size. If the representation of the Relative Network Size from the other network's beacon is larger than node 200's current copy of the Relative Network Size, then a network merge is required. If the representation of the Relative Network Size from the other network's beacon is smaller than node 200's current copy of the Relative Network Size, then node 200 will discard the information received in the beacon from the other network and will remain in synchronization maintenance mode looking for beacons with the same BSSID. During initial synchronization and synchronization maintenance, each node adds its own bit to the other bits in the bit map received in a beacon (initially all bits are set to zero) provided that the Network Size Sequence Number received in the beacon is larger than the Network Size Sequence Number last broadcast in a beacon by node 200. The Network Size Sequence Number is a beacon parameter that accompanies the Relative Network Size beacon parameter. Once node 200 adds a bit to the Relative Network Size, it increments its own copy of the Network Size Sequence Number before broadcasting the next beacon.
Finally, a periodic reset of the Relative Network Size bit map is recommended during synchronization maintenance. Any node can initiate this reset. A node will reset the Relative Network Size bit map (thus initializing the network size representation), set its own bit, and then increment the beacon sequence number. Once a surrounding node hears the beacon with an incremented sequence number, it knows that it should add its bit to the map and propagate the new bit map in subsequent beacons that it broadcasts.
Synchronization Procedure (Partial Network Coordinator Coverage Available)
When a node moves in and out of coverage of a network coordinator, three possible scenarios exist for synchronization and association. The node is either, a) unsynchronized with any network, b) synchronized with a non-network coordinator reference clock, or c) synchronized with a network coordinator reference clock. These 3 scenarios will be addressed below.
Merging Unsynchronized Node into a Mesh Service Area
This first scenario does not require any change from the current method of association and synchronization of a node with a network coordinator (i.e. becoming a mesh member of a network coordinator). This occurs with mobility or power on scenarios and only applies to single hop synchronization. Once node 200 becomes a mesh member of the network coordinator, it will set its Synchronization Tier Number to 1 (the network coordinator will always be a tier #0 node) and the Infrastructure Access Index indication will be set to 1.
Merging IBSS Synchronized node into Mesh Membership with a Network Coordinator
This scenario represents the transition from a disconnected mesh to a connected mesh. This occurs when a node is synchronized with other nodes outside of mesh membership with a network coordinator, and is now able to receive beacons from a network coordinator (i.e. IBSS synchronized nodes desire to synchronize with network coordinator). These IBSS synchronized nodes may be sleeping in power save mode. Since it is likely that their beacons are being broadcast asynchronously to the network coordinator beacons, every TBSS
Dropping Mesh Membership with a Network Coordinator
This scenario represents the possible transition from a connected mesh to a disconnected mesh. This occurs if a tier #1 node with mesh membership to a network coordinator stops receiving beacons from tier #0 (i.e. the network coordinator). This scenario has two possible outcomes:
Once a node stops receiving beacons from any lower, equal, or higher tier nodes with the Infrastructure Access Index indication set to 1, then the WLAN mesh is entering a disconnected mesh state. At this time, node 200 can start a timer Tresync
While the invention has been particularly shown and described with reference to a particular embodiment, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention. It is intended that such changes come within the scope of the following claims.